Anim. Behav., 1989,37,498-506
Does dominance determine how far dark-eyed juncos, Junco hyemulis,
migrate into their winter range?
CHRISTOPHER
Department
M. ROGERS,*
TAD L. THEIMER,?
ELLEN
D. KETTERSON
qf Biology, Indiana University, Bloomington,
VAL
NOLAN
JR &
Indiana 47405, U.S.A.
Abstract. The behavioural dominance hypothesis suggests that differential migration among individuals
of a species of bird is due solely to social interactions that force subordinate individuals (often a class, e.g.
female or young birds) to migrate farther into the winter range than dominant individuals (often a class,
e.g. male or old birds). Here, this hypothesis was tested with two experiments. In the first experiment, the
prediction was tested that dominance acts within a sex-age class and influences migration distance. The
outcomes of interactions within dyads of dark-eyed juncos, Junco hyemalis, were observed. Each dyad
consisted of a junco caught in winter in Michigan matched with another of the same sex-age class caught
in Indiana, which is situated farther south and therefore farther into the winter range of this species.
Michigan birds were dominant in only half of the experimental
dyads (21 of 41 dyads), which is
inconsistent with the prediction of the dominance model. In the second experiment the prediction was
tested that members of a sex-age class that migrates farther into the winter range should be subordinate to
members of a different class that migrates a shorter distance. Young males that wintered in Michigan were
pitted against old males that wintered in Indiana. In 19 of 25 dyads, the more southern-wintering
old males
were dominant, which also is counter to the prediction of the dominance hypothesis. These results
indicate, at the very least, that in migratory J. h. hyemalis, dominance does not play as important a role in
determining latitude of winter residence as has been suggested.
Migration
is one form of behaviour
by which
animals that breed at temperate latitudes may
escape the rigours of winter. Differential migration
in temperate-breeding
birds occurs when individuals of the same species, but different sex-age
classes, migrate varying distances into the wintering grounds. Differential
migration
of sex-age
classes has been found in a wide variety of avian
species (reviewed by Ketterson & Nolan 1976,
1983, 1985; Myers 1981; Gauthreaux
1982). In
general, females and young overwinter farther into
the wintering range than males and adults.
Various hypotheses have been advanced to
explain such intraspecific variation in avian migration (for review see Gauthreaux 1982; Ketterson &
Nolan 1983). The behavioural
dominance hypothesis (Gauthreaux
1978, 1982) states that in social
competition
for resources, the sex-age classes
experience unequal success because they differ in
* Present address: Department of Zoology, University of
British Columbia, 6270 University Boulevard, Vancouver, British Columbia V6T 2A9, Canada.
t Present address: Department of Biological Sciences,
Northern Arizona University, Flagstaff, Arizona
8601 I, U.S.A.
0003-3472/89/030498
+09$03.00/O
social dominance status. According to the hypothesis, if resources on the breeding grounds are
inadequate for an extended period such as a winter
season, then dominant birds will travel only as far
into the winter range as is necessary to find
sufficient food, and subordinate birds will move
farther in order to avoid competition
with the
dominant
birds. The result predicted
by the
hypothesis is a linear relationship between potential dominance status and distance migrated into
the winter range. No studies to date have directly
tested this prediction of the dominance hypothesis.
In this study, we apply the dominance model to a
bird species in which individuals migrate north to
south between virtually non-overlapping
breeding
and wintering ranges, and settle for the winter over
a considerable band of latitudes. In this situation,
the model asserts that individuals
wintering
at
latitudes closer to the breeding range should be
dominant over those wintering at greater distances.
While the entire distance migrated by any individual typically cannot be known merely from its
winter location, because its breeding origin is
unknown, the distance migrated into the winter
range can be measured. In species with non-
((:I989 The Association for the Study of Animal Behaviour
498
Rogers et al.: Dominance
overlapping
seasonal ranges, only this distance
(and not the actual distance migrated) has the
potential to determine with which conspecifics an
individual
will overwinter.
Therefore
it is the
distance that is critical to the dominance hypothesis
(Gauthreaux
1978, 1982).
Our experimental subject, the dark-eyed junco,
Junco h. hyemalis, is a migratory bird species that
breeds principally in Canada and spends the winter
southward in southern Ontario and most of the
eastern United States (Bent 1968); the breeding and
wintering ranges overlap only slightly. Juncos are
gregarious during winter, and, when in flocks,
males tend to dominate females and old birds tend
to dominate young birds (Balph 1977; Baker & Fox
1978; Ketterson 1979a). On average, males winter
north of females (Ketterson & Nolan 1976, 1979)
and young of the year winter north of adults of the
same sex (Ketterson & Nolan 1983, 1985).
In the first of two experiments, we tested the
prediction that in dyads of juncos of the same sexage class, the member from the more northern
wintering site should dominate the member from
the more southern site. The second experiment
tested the same prediction by measuring inter-class
differences in social dominance. Even though old
males dominate
young males when they are
members of the same population
(see above), the
dominance
hypothesis would predict that the
opposite would be true of young males that winter
north of old males. We therefore determined
dominance status in junco dyads each consisting of
a northern-wintering
young male and a southernwintering old male. Because previous studies (Ketterson & Nolan 1982, 1983) have shown some
annual variation in the age structure of winter
populations
of juncos at various latitudes, we
sampled age structure in the winter of the second
experiment in order to learn whether we were
performing
the experiment
in a normal or an
aberrant year.
METHODS
Experimental Junco Populations
In the junco, fall migration ends by approximately 1 December throughout eastern North America (Ketterson &Nolan 1976, 1985), and thereafter
populations are considered stable until approximately 1 March, when spring migration begins (Ketterson & Nolan 1976, 1985; but see Terrill 1987,
and migration
in juncos
499
who proposes that facultative migration may occur
in mid-winter if feeding conditions deteriorate). All
birds used in the present study were selected from
two winter populations,
one near Bloomington,
Indiana (39”N latitude) and one 350 km northward, near Kalamazoo, Michigan (42”N latitude).
Michigan (MI) experimental birds were captured at
the Kalamazoo
Nature Center, a large nature
preserve including extensive forest edge habitat
occupied by wintering juncos. Indiana (IN) juncos
were taken from similar habitat at two locations.
Experiment 1
This experiment
addressed the question of
whether dominance determines the winter site in
juncos of the same sex and age.
Capture dates
Birds were caught in two winters, as follows. In
1984-1985,
birds were captured
on 15-19
December 1984 (young males and young females in
Indiana),
15-25 February
1985 (old males in
Indiana), and 20-22 December 1984 (old males,
young males, young females in Michigan). In 19851986, birds were captured on l-l 1 January 1986
(young females in Indiana) and 6-10 January 1986
(young females in Michigan).
Old females are
uncommon as far north as Michigan and Indiana
and were not included in the experiment.
Sex and age determination
Birds were sexed on the basis of plumage traits
and wing length, after Ketterson & Nolan (1976).
Two age categories are recognizable in this species;
young juncos are those hatched the preceding
breeding season, old juncos are in at least their
second winter. Age determination
by inspection of
skull ossification is reliable until approximately
1
January (Ketterson & Nolan 1982), and we relied
on this character until that date. Thereater we used
a combination
of plumage traits, wing length and
eye colour that has proved 93% accurate (Ketterson & Nolan 1982).
Dyad establishment and housing conditions
After capture and until determination
of dominance status, juncos from different winter populations were kept visually isolated in two large
separate aviaries at Bloomington,
Indiana. Food
and water were provided ad libitum, population
densities were moderate, and free flight in the
aviaries was possible.
500
Animal
Table
1)
I. Aspects
TYWof
&ad
Young
of individual
Dvad
number
Winner
dyads
including
No. of
displacements
number
37, 3
of displacements
No. of
avoidances
and number
Wing length (mm)
Winner-loser
of avoidances
(experiment
Body mass (g)
Winner-loser
Fat class
Winner-loser
-.-.-.___
males
1
2
Winner-loser
Young
junco
Behaviour,
(R+
IO
II
12
13
14
15
16
1 SE)
Michigan
Michigan
Michigan
Indiana
Michigan
India&
Michigan
Indiana
Indiana
Michigan
Indiana
Indiana
Indiana
Michigan
Michigan
Indiana
15
11
15
15
4
24
16
14
21
19
21
19
20
14
15
20
0
0
0
0
3
13
0
I
5
4
5
8
8
3
3
2
Michigan
Michigan
Michigan
Michigan
Michigan
Michigan
India;
Indiana
Indiana
Michigan
Michigan
Indiana
Indiana
Michigan
Michigan
Indiana
Indiana
Indiana
Michigan
Michigan
Michigan
19
15
16
19
15
20
2.5
5
15
27
12
20
15
69
3
15
12
16
9
13
7
0
0
0
I
0
2
0
1
5
6
1
0
3
1
0
0
0
5
0
0
0
0
0
0
-1
0
0
0
0
0
0
0
0
0
0
-1
0.06+0.11
-0.2
-0.6
I.1
0.8
- I.5
2.5
-2.5
0
- 1.0
-0.9
-0.6
I.3
0.1
1.8
-0.8
-0.1
0.04+0.32
0
0
-0.5
1,o
- I.5
- I.0
0
- I.0
0.5
0.5
0.5
0
-15
0.5
-0,18+0.22
females
2
3
4
6
Winner-loser
(X*
8
9
10
II
12
13
14
15
16
17
18
19
20
21
Isa)
0
0
0
0
-1
-1
-1
0~05+0~13
-0.4
I.3
0.4
0
I.8
-1.0
2.0
2.5
0.7
- I.3
- I.7
- I.1
1 .o
-0.5
2.2
-0.5
0.9
-0.9
0.8
0.2
0.1
0.3 1 + 0.27
0
0
0
I.0
0
0.5
0.5
I.0
I.0
-0.5
-0.5
0
0
-0.5
0
-0.5
- I.0
- I.0
I.0
-0.5
0.5
0~07+0.13
-4.2
-1.0
I.5
0.1
-0,75+
I.46
- I.0
-0.5
0.5
0
-0.25+0.37
Old males
4
Winner-loser
(Z+
1
SE)
Indiana
Indiana
Indiana
Indiana
15
15
15
15
2
1
1
0
-1
0
0.25 * 0.55
The quantity
winner-loser
refers to the difference
between, for example, the wing lengths of the contest winner and
loser; such quantities
are also shown for body mass and fat class. Average (+ 1 SE) winner-loser
values are indicated
for separate age-sex classes.
Rogers
et al.:
Dominance
and migration
501
in juncos
Table II. Aspects of individual junco dyads including number of displacements and number of
avoidances (experiment 2)*
Young males versus old males
Dyad
number
I
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Winner
Indiana
Michigan
Indiana
Indiana
Indiana
Indiana
Indiana
Indiana
Indiana
Indiana
Indiana
Michigan
Indiana
Indiana
Indiana
Michigan
Michigan
Indiana
Indiana
Michigan
Indiana
Michigan
Indiana
Indiana
Indiana
No. of
displacements
12
31
15
15
21
19
17
16
15
16
17
38
15
11
26
7
14
16
15
34
24
15
20
No. of
Wing length (mm) Body mass (g)
Fat class
avoidances
Winner-loser
Winner-loser
Winner-loser
0
1
18
0
8
0
0
6
3
2
5
-I
-1
-2
2
4
-1
4
-4
3
4
3
I.70 + 0.48
0
0.2
-1.2
-3.4
1.3
- I.5
0.1
-0.7
-0.8
1.7
0.1
0.5
0.6
-0.24*0.37
0.5
-0.5
0
-0.5
o-5
- 1.0
- 1.0
1.0
-1.0
0.5
I.0
0.5
0
* For a complete explanation, see Table I.
To form dyads, MI juncos were paired with IN
juncos of the same age, sex, wing length (flattened
wing, nearest 1.O mm) and body mass (Pesola scale,
nearest 0.1 g), as measured at the time of pairing.
These factors were held constant because they have
been shown to be correlated with dominance status
in this species (Fretwell 1969; Balph 1977; Baker &
Fox 1978; Ketterson 1979a). Thus, members of
dyads were similar in all but one important respect:
they terminated
their fall migration
at winter
latitudes that were 350 km apart. We emphasize
that there is no geographical
variation in wing
length (a general measure of body size, discussed by
James 1970) within sex-age classes among winter
populations
of this species (Nolan & Ketterson
1983). Therefore, when we matched members of
dyads for wing length, we did not eliminate naturally occurring
geographical
variation
in size.
Finally, by pairing birds of similar visible subcuta-
neous fat class (determined on a scale of O-5, after
Helms & Drury 1960; Nolan & Ketterson 1983) we
attempted to control for the influence of stored
energy on the motivation to feed in captivity at the
time dyads were caged and dominance status was
established. In sum, our direct control allowed a
clear focus on the factor of major concern: chosen
latitude of winter residence.
During observations, dyads were housed separately in cages (0.6x0.6 x0.6 m) made of black
plastic (back, bottom, two sides) and hardware
cloth (front and top). Individual dyads could hear
but not see dyads in other cages. Two parallel
perches, 0.2 m apart and 0.3 m high, extended from
the back of each cage. Ten (1984-1985) and 14
(1985- 1986) identical cages allowed observation of
several dyads at once. White millet and cracked
corn, scattered over the entire cage floor, and water
(snow in cold conditions) were provided ad libitum
502
Animul
Behaviour.
37, 3
60
i
Young males
Old ;noles
Young females
Old females
Figure 1. Relative
abundances
of the four junco age--sex classes at Kalamazoo,
Michigan
(m, 42’ N, N=255)
and
Bloomington,
Indiana (0,39”N,
N=485)
in December-early
January
198551986. Absolute
frequencies
of males were:
MI = 130 young, 39 old; IN = 187 young, 163 old. A test ofthe 2 x 2 contigency
table (MI, IN versus young, old juncos)
for males was highly significant
(x2= 26.47, df= 1 PC O,OOl), indicating
that old males, on average, selected a more
southerly
latitude for overwintering
than young males.
at all times.
A constant
light regime (1984.-1985:
incandescent
bulbs in a windowless room) of
1ight:dark 9: 15-h or a natural light regime (19851986: cages held outdoors) was provided. The
experimental cages were on the second floor of an
unheated building in 1984-1985 and just outside
the building in 1985-1986; thus experimental birds
experienced natural temperature fluctuations typical of southern Indiana in winter.
Criteria used to determine dominance status
We defined two categories of behaviour that
indicated dominance of one member of a dyad over
the other. Displacement consisted of a direct attack
upon one bird by the other, which always resulted
in rapid retreat by the attacked bird and occupation of its space by the attacker. Conflicts of this
type occurred most often over food and perches but
occasionally arose over water or snow. Avoidance,
a more subtle behaviour, involved clear movement
by one bird away from the other as the latter moved
about the cage. Almost all avoidances occurred
when one bird was feeding and hopped in the
general direction of the second bird, also feeding,
which moved aside.
A bird seen to displace its cagemate 15 times was
classified as dominant. In a few cases (11 of 41
dyads) displacements were uncommon (fewer than
15) and we relied upon avoidances as well as
displacements to make our determinations.
We
emphasize, however, that we observed no reversals
in either displacement or avoidance behaviour in
any dyad in this experiment (or in the one to
follow), suggesting highly stable dominance relations between dyad members. This is in accordance
with previous studies on this species (Balph 1979;
Ketterson 1979a).
Dyad members were introduced simultaneously
into the observation cages in order to control for a
possible effect of prior residence (Balph 1979;
Yasukawa & Bick 1983). A total of 10 cages was
used at once to house dyads, so it was necessary to
determine dominance status in successive shifts in
order to accumulate sample size. Groups of 10 or
fewer dyads were established on 30 December 1984
(five of young males and five of young females),
9 January 1985 (six of young males and three of
young females), 15 January 1985 (five of young
males and five of young females), 26 February 1985
(four of old males), and 14 March 1986 (eight of
young females). Observation began 1-8 days after
establishment
(32/41 dyads were first observed
after at least 2 days had passed, and the mean was
3.6 days) and continued until dominance status had
been determined. On average, we observed each
dyad during 2.6 days (range 14 days).
Experiment
2
This experiment
addressed
whether dominance determines
juncos of different age.
the question of
the winter site in
Rogers et al.: Dominance
Capture dates; sex and age determination
This experiment was conducted in the winter of
19851986. Capture dates of old and young males
were l-11 January 1986 in Indiana, and 6-10
January 1986 in Michigan.
Sexing and ageing
proceeded as in experiment 1.
Dyad establishment
Prior to dominance testing, MI and IN birds
were housed separately as described for experiment
1. Because dyad members differed in age class in
experiment 2, we did not control for age-associated
factors affecting dominance status, i.e. wing length
and, correlated with wing length, body size (both
are significantly greater in old than young males,
Nolan & Ketterson
1983). Experimental
dyads
were housed outdoors as described above for the
1986 dyads employed in experiment 1.
Dyad observation
We again used displacements and avoidances to
determine relative dominance status of the dyad
members. As in experiment
1, members were
introduced simultaneously
to experimental cages.
Groups of old male-young male dyads were established on 26 January 1986 (12 dyads) and 7
February 1986 (13 dyads). Observation began 3-5
days after establishment
(23/25 dyads were first
observed after 3 days had passed, and the mean was
3.2 days) and continued until dominance status had
been determined. On average, we observed each
dyad during 1.7 days (extremes l-4 days).
Sampling of 1986 winter junco populations
Sampling
of natural junco populations
at
Bloomington,
IN took place from 4 to 31
December 1985, and 485 juncos were caught, aged
and sexed. A comparable effort at Kalamazoo, MI
from 6 to 10 January 1986 yielded 255 juncos.
RESULTS
Experiment 1
Latitude of winter residence was not significantly associated with dominance status (Table I).
Of the 16 contests involving young males, eight
were won by IN juncos. In the interactions of 21
young female dyads, Indiana females were dominant in eight. In the adult male group, all four
contests were won by IN juncos. After pooling, 20
of the 41 contests were won by IN and 21 by MI
juncos; this distribution
did not differ significantly
from a 50: 50 win: lose ratio (x2 =0.024, df= 1,
and migration
in juncos
503
P > 0.90). When outcomes based on fewer than 15
displacements (Table I: four dyads of young males
and six of young females) were eliminated from
analysis, the statistical result was unchanged
(x2=0.290, df= 1, P>O.5).
Because we matched dyad members for potential
determinants of dominance, it is not surprising that
winners and losers did not differ in wing length,
body mass, or fat class (t-tests of wing length and
body mass, Mann-Whitney
U-test of fat class,
Table I). Thus the only known difference remaining
between them, chosen wintering
site, did not
influence dominance status.
Experiment 2
In marked contrast to the first experiment, in
experiment 2 the latitude of winter residence was
significantly associated with dominance status, but
in a direction opposite to that predicted by the
dominance hypothesis (Table II). Of the 25 contests involving old IN males versus young MI
males, 19 were won by the old IN males (Table II;
x2=6.760, df=l,
P<O.Ol,). Again, when dyads
with fewer than 15 displacements were eliminated,
the result was unchanged
(x2 =4.260,
df= 1,
P<O.O5). Interestingly, ties occurred in several of
these mixed-age
dyads when a flying bird
attempted to displace a perched bird but failed.
These were rare, however, and even if they had been
treated as reversals would not have contradicted
assigned dominance status.
Sampling of the natural MI and IN winter
populations
showed that, as in previous years
(Ketterson & Nolan 1983) in the winter of experiment 2 old males formed a significantly
higher
proportion of the total junco population in Indiana
than in Michigan (see Fig. 1). That is, there was
nothing unusual about the winter demography
when experiment 2 was conducted, and there was
no reason to suppose that its subjects were drawn
from populations with atypical structures.
DISCUSSION
Implications
Migration
for Dominance-influenced
Differential
Proponents of the dominance hypothesis have
advanced it to account both for segregation or
partial segregation of the sexes and for differences
in the distribution
of age classes. Since migrant
504
juncos exhibit
our discussion
sex.
Animul Behaviour,
both forms of winter segregation,
is separated according to age and
Age
The results of experiment 2 indicate that old
males from IN dominated young males from Ml.
The novel element of this experiment is that the
birds were taken from two latitudes; those from the
more southerly latitude were dominant,
which
contradicts
the prediction
of the dominance
hypothesis. Furthermore,
the contemporaneous
sampling of the two natural populations
from
which the experimental subjects were drawn shows
that young birds, as usual (Ketterson & Nolan
1983), tended to settle north of old birds. Can the
experimental result be explained as an artefact of
our methods? We can think of three possibilities.
First, contests in experiment 2 were held in a
neutral and unnatural
setting, which may have
deprived the young juncos of an advantage that
they would have had in nature. Contests in cages
are likely to be dependent on larger body size,
greater experience (age), and darker plumage; old
males should dominate under these circumstances
(Ketterson
1979b). If young males arrive in the
northern part of the winter range before old males,
prior residence (Balph 1979; Ketterson
1979a;
Yasukawa & Bick 1983; Ketterson & Nolan 1985)
might provide them with an advantage that would
outweigh the advantages inherent in greater age
and permit young juncos to dominate in natural
contests during migration.
The evidence as to
whether young males arrive before old is mixed. In
an autumn Zugunruhe experiment, young of the
year showed readiness to migrate at an earlier date
than did adults (Ketterson & Nolan 1985). Also,
when young juncos are banded in Indiana and
return there the following year, the date of capture
tends to be later in the second winter than in the
first winter (Ketterson & Nolan 1985). In contrast,
there is a high proportion
of old males among the
earliest autumn
migrant
juncos
in Indiana
(October, early November), higher than the proportion found a few weeks later after the winter
population has settled (Nolan & Ketterson, unpublished data). Thus, at least some old males migrate
very early. In sum, we cannot exclude the possibility that young Michigan juncos would have dominated old Indiana juncos under more natural
circumstances, but at least on neutral ground they
do not.
37, 3
A second factor that might have led us to a false
conclusion
is that two study populations
had
inadvertently been selected from habitats of different quality. In addition to its prediction of geographical segregation, the dominance hypothesis
predicts that when microgeographical
variation in
habitat quality exists at a single location and
superior habitat is in short supply, subordinates
will be forced into habitats of lower quality
(Gauthreaux 1978). If this prediction is correct and
if young MI males were obtained from low-quality
habitat into which they had been forced by other,
more dominant young MI males, we could have
dealt with an unrepresentative sample of MI birds.
However, in both Michigan and Indiana juncos
were caught in habitats that appeared to be
comparable and of high quality. In both locations,
after I December natural food was supplemented
and in both locations the presence of coniferous
trees provided
numerous,
protected
potential
roosts. Hence, we have no reason to believe our
subjects were atypical members of their respective
classes.
Finally, it may be argued that our results can be
reconciled with the dominance hypothesis on the
ground that our Michigan and Indiana sites were
too close to each other ‘and that therefore no
correlation
between latitude and dominance relations could be expected. This too seems to us to be
an unlikely explanation.
Juncos are abundant in
winter in the eastern United States only between
33 and 42’N (Ketterson & Nolan 1983, 1985). Our
sites were separated by three degrees of latitude,
thus 25% of the region of abundance. Furthermore, it is within this 25% that over half the young
male population
winters. The argument that the
dominance hypothesis is not designed to predict
distributions
on this scale is difficult to accept.
Neither experiment bears directly on the adequacy of the dominance hypothesis to account for
the junco’s sexual distribution
in winter, and we
think it possible that an ultimate cause of the longer
migrations of females may be the advantage of
reduced competition
with males during winter
(Ketterson
& Nolan 1983). Social interactions
during autumn migration may also play a proximate role in determining
where females settle.
Nevertheless, the evidence that dominance explains
neither within-class nor across-male-age-class dis-
Rogers
et al.:
Dominance
tributions
suggests one must be cautious before
accepting the dominance hypothesis with respect to
sex.
Juncos
in general
While the distribution
of eastern migratory
juncos violates the dominance model’s straightforward prediction that dominant birds (males and
adults) should winter north of subordinate birds
(females and young), factors underlying the migratory junco’s distribution
may not be typical (compare Rabenold & Rabenold 1985; Wiedenmann &
Rabenold 1987 for inter-racial comparisons). The
foundation of the dominance model is the assumption that in competition for scarce resources subordinates must either emigrate or face the probability
of reduced survivorship. There is evidence for this
assumption
in some species. Thus, artificially
increased food supplies raised winter survivorship
in subordinate song sparrows (Melospiza
melodia;
Smith et al. 1980) and dominant silvereyes, Zosterops lateralis,
survived better in winter than subordinates (Kikkawa 1980). In contrast, demographic
data on juncos that winter in Indiana and South
Carolina suggest that neither overwinter survival
nor winter dispersal varies according to dominance
status of age-sex classes. When large numbers of
juncos were aged, sexed, banded and released at the
beginning of winter and the banding sites sampled
again at winter’s end (with no artificial
food
supplied in the interim), two findings emerged. (1)
The proportion
of banded individuals recaptured
at the end of winter was invariant across age-sex
classes, and (2) the sex-age structure of the earlywinter and late-winter samples (i.e. counting all
juncos, banded and unbanded,
in late winter)
remained unchanged (Ketterson & Nolan 1983,
1985, additional
unpublished data). It is possible
that the junco’s winter distribution
may be idealfree as opposed to despotic. That is, fitness may be
independent
of potential
dominance
status or
wintering site.
A recent study by Terrill (1987) provides an
interesting contrast. In his experiments, after dominance had been established between members of
dyads, food was severely restricted and the amount
of nocturnal restlessness of dyad members was then
measured. (Nocturnal restlessness accompanied by
fattening during the seasons of migration is generally accepted as evidence that caged migratory
birds, including juncos, are in the migratory physiological state; Berthold 1975; Ketterson & Nolan
and migration
in juncos
505
1987a, b.) Subordinate juncos became-more restless
than dominant juncos, and in one experiment their
movements were oriented southward. None of the
birds fattened. Terrill concluded that subordinate
juncos should migrate out of an area before
dominant juncos ‘in situations where dominants
are able to maintain relatively high probabilities of
survival, but directly decrease the probability
of
survival by subordinates’.
Our results would suggest that a similar situation does not occur in the
course of a typical junco fall migration.
The results of these experiments suggest that for
the junco, the dominance
hypothesis must be
modified to admit the possibility that additional
selective pressures underlie the differential migration and winter distribution.
Ketterson & Nolan
(1983) have proposed other pressures, which may
bear unequally on the age-sex classes; optimal
balancing of these factors by members of each class
might then result in the observed winter distribution. Among the suggestions are: (1) the relationships among proximity to the breeding range, early
return to the breeding range, and reproductive
success (perhaps more important
for males than
females); (2) mortality rate per unit distance traveiled during migration (possibly higher in young
birds with no previous experience ofmigrating);
(3)
achievement of a compensatory trade-off between
greater mortality associated with longer migration
but lower overwinter
mortality associated with
wintering in a more southerly climate (possibly
more attainable
by older birds that are more
experienced at avoiding the risks inherent to migration). Finally, to the extent that these pressures
make one class more resistant than others to
prolonging
its southward migration, those other
classes might then move farther southward to
avoid the consequences of high population density.
This could account for the longer migration
of
adults than of young birds of the same sex.
ACKNOWLEDGMENTS
We thank Kitty Gehring and Krissy Brunner for
generous help with the experimental
set-up and
maintenance,
respectively. Ray Adams and the
Kalamazoo
Nature Center kindly made capture
opportunities
possible in Michigan.
Ken Jones,
Robin Jung and Jim Kaiser were very helpful field
assistants. This manuscript benefited substantially
from comments made by K. Able, M. Baker, M.
Animul
506
Beha Gxir.
Balph, P. Ficken, S. Fretwell, B. Hulbertson,
J.
Kikkawa,
J. Smith, S. Terrill and R. Walton.
Helpful discussions were also had with S. Gauthreaux, H. R. Pulliam, J. P. Myers, K. Rabenold and
S. Rohwer. This study was supported by Indiana
Academy of Science and Sigma Xi grants to CMR,
and by NSF grant BNS-83- 15348 to EDK and VN.
REFERENCES
Baker, M. C. &Fox,
S. F. 1978. Dominance,
survival,
and
enzyme
polymorphism
in dark-eyed
juncos,
Junco
hyemalis.
Evolution,
32, 697-711.
Balph, M. H. 1977. Winter social behaviour
of dark-eyed
juncos:
communication,
social organization,
and eco.
logical implications.
Anim. Behav., 25, 859-884.
Balph, M. H. 1979. Flock stability
in relation
to social
dominance
and agonistic
behavior
in wintering
darkeyed juncos. Auk, 96,714-727.
Bent, A. C. 1986. Life Histories
of North
American
Cardinals,
Grosbeaks,
Buntings,
Towhees,
Finches,
Sparrows,
and Allies. New York: Dover.
Berthold,
P. 1975. Migration:
control
and metabolic
physiology.
In: Avian Biology.
Vol. 5 (Ed. by D. S.
Farner & J. R. King), pp. 77-128.
New York:
Academic Press.
Fretwell,
S. D. 1969. Dominance
behavior
and winter
habitat distribution
in juncos (Junco hyemalis).
BirdBanding, 40, l-25.
Gauthreaux,
S. A., Jr. 1978. The ecological
significance
of
behavioral
dominance.
In: Perspectives
in Ethology
(Ed. by P. P. G. Bateson & P. H. Klopfer),
pp. 17-54.
New York: Plenum Press.
Gauthreaux,
S. A., Jr. 1982. The ecology and evolution
of
avian migration
systems. In: Auiun Biology.
Vol. 6 (Ed.
by J. R. King & D.S. Farner),
pp. 93-167. New York:
Academic
Press.
Helms,
C. W. & Drury,
W. H. 1960. Winter
and
migratory
weight and fat: field studies on some North
American
buntings.
Bird-Bonding,
31, I-40.
James, F. C. 1970. Geographic
size variation
in birds and
its relationship
to climate. Ecology, 51, 3655390.
Ketterson,
E. D. 1979a. Aggressive
behavior
in wintering
dark-eyed
juncos: determinants
ofdominance
and their
possible relation to geographical
variation
in sex ratio.
Wilson Bull., 91, 371-383.
Ketterson,
E. D. 1979b. Status signaling in juncos. Auk,
96, 94-m99.
Ketterson,
E. D. & Nolan,
V., Jr. 1976. Geographic
variation
and its climatic
correlates
in the sex ratio of
eastern-wintering
dark-eyed
juncos
(Junco hyemalis
hyemalis).
Ecology,
57,679-693.
Ketterson,
E. D. & Nolan, V., Jr. 1979. Seasonal, annual,
37, 3
and geographic
variation
in sex ratio of wintering
populations
ofdark-eyed
juncos (Junco hJaema/is). Auk,
96, 532-536.
Ketterson,
E. D. & Nolan,
V., Jr 1982. The role of
migration
and winter mortality
in the life-history
of a
temperate-zone
migrant,
the dark-eyed
junco. as determined from demographic
analyses
of winter populations. Auk, 99, 243-259.
Ketterson,
E. D. & Nolan, V.. Jr. 1983. The evolution
of
differential
bird migration.
In: Currenr
Ornithologic.
Vol. I (Ed. by R. F. Johnston),
pp. 357-401. New York:
Plenum Press,
Ketterson,
E. D. & Nolan,
V., Jr. 1985. Intraspecific
variation
in avian migration:
evolutionary
and regulatory aspects. In: Migration:
Mechanisms
and Adaptive
SigniJicance
(Ed. by M. A. Rankin),
pp. 553-579. Port
Aransas,
Texas: University
of Texas.
Ketterson,
E. D. & Nolan,
V., Jr. 1987a. Spring and
summer
confinement
of dark-eyed
juncos at autumn
migratory
designation
suppresses
normal
autumn
behaviour.
Anim. Behav.. 35, 1744-1753.
Ketterson.
E. D. & Nolan, V., Jr. 1987b. Suppression
of
autumnal
migratory
unrest in dark-eyed
juncos held
during summer
on, near, or far from their previous
wintering
sites. Auk, 104, 303 3 10.
Kikkawa,
J. 1980. Winter
survival
in relation
to dominance classes among
silvereyes
Zosterops
lateralis
chlorocephala
of Heron
Island,
Great Barrier
Reef.
Ibis, 122,437 -446.
Myers,
J. P. 1981. A test of three hypotheses
for
latitudinal
segregation
in wintering
birds. Can. J. Zool..
59, 1527m 1534.
Nolan, V.. Jr & Ketterson,
E. D. 1983. An analysis
of
body mass, wing length, and visible fat deposits
of
dark-eyed
juncos wintering
at different
latitudes.
Wilson Bull., 95, 603-620.
Rabenold,
K. N. & Rabenold,
P. P. 1985. Variation
in
altitudinal
migration,
winter
segregation,
and site
tenacity
in two subspecies
of dark-eyed
juncos
in the
southern
Appalachians.
Auk, 102,8055819.
Smith, J. N. M.. Mongomerie,
R. D. & Yom-Tov,
Y.
1980. A winter feeding experiment
on an island song
sparrow
population.
Oecologia
(Berl.),
47, 164-170.
Terrill,
S. 8. 1987. Social dominance
and migratory
restlessness
in the dark-eyed
iunco, Junco hyrmali,y.
Behac. Ecol. Sociobiol.,
21; 1 il.
Wiedenmann.
R. R. & Rabenold.
K. N. 1987. The effects
of social dominance
between two subspecies
of darkeyed juncos (Junco hyemalis).
Anim. Behav., 35, 856864.
Yasukawa,
K. & Bick, E. I. 1983. Dominance
hierarchies
in dark-eyed
juncos (Junco hyemalis):
a test of a game
theory model. Anim. Behrn., 31,439-448.
i Received
7 July 1987; revised I3 June
M.S. number: A4930)
IYX(I,